<feed xmlns='http://www.w3.org/2005/Atom'>
<title>user/sven/linux.git/include/uapi/linux/fuse.h, branch v3.10.88</title>
<subtitle>Linux Kernel
</subtitle>
<id>https://git.stealer.net/cgit.cgi/user/sven/linux.git/atom?h=v3.10.88</id>
<link rel='self' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/atom?h=v3.10.88'/>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/'/>
<updated>2013-05-01T12:37:21Z</updated>
<entry>
<title>fuse: add flag to turn on async direct IO</title>
<updated>2013-05-01T12:37:21Z</updated>
<author>
<name>Miklos Szeredi</name>
<email>mszeredi@suse.cz</email>
</author>
<published>2013-05-01T12:37:21Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=60b9df7a54804a965850db00beec4d3a2c002536'/>
<id>urn:sha1:60b9df7a54804a965850db00beec4d3a2c002536</id>
<content type='text'>
Without async DIO write requests to a single file were always serialized.
With async DIO that's no longer the case.

So don't turn on async DIO by default for fear of breaking backward
compatibility.

Signed-off-by: Miklos Szeredi &lt;mszeredi@suse.cz&gt;
</content>
</entry>
<entry>
<title>fuse: fix type definitions in uapi header</title>
<updated>2013-04-17T10:30:40Z</updated>
<author>
<name>Miklos Szeredi</name>
<email>mszeredi@suse.cz</email>
</author>
<published>2013-04-17T10:30:40Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=4c82456eeb4da081dd63dc69e91aa6deabd29e03'/>
<id>urn:sha1:4c82456eeb4da081dd63dc69e91aa6deabd29e03</id>
<content type='text'>
Commit 7e98d53086d18c877cb44e9065219335184024de (Synchronize fuse header with
one used in library) added #ifdef __linux__ around defines if it is not set.
The kernel build is self-contained and can be built on non-Linux toolchains.
After the mentioned commit builds on non-Linux toolchains will try to include
stdint.h and fail due to -nostdinc, and then fail with a bunch of undefined type
errors.

Fix by checking for __KERNEL__ instead of __linux__ and using the standard int
types instead of the linux specific ones.

Reported-by: Arve Hjønnevåg &lt;arve@android.com&gt;
Reported-by: Colin Cross &lt;ccross@android.com&gt;
Signed-off-by: Miklos Szeredi &lt;mszeredi@suse.cz&gt;
</content>
</entry>
<entry>
<title>fuse: allow control of adaptive readdirplus use</title>
<updated>2013-02-07T13:25:44Z</updated>
<author>
<name>Eric Wong</name>
<email>normalperson@yhbt.net</email>
</author>
<published>2013-02-06T22:29:01Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=634734b63ac39e137a1c623ba74f3e062b6577db'/>
<id>urn:sha1:634734b63ac39e137a1c623ba74f3e062b6577db</id>
<content type='text'>
For some filesystems (e.g. GlusterFS), the cost of performing a
normal readdir and readdirplus are identical.  Since adaptively
using readdirplus has no benefit for those systems, give
users/filesystems the option to control adaptive readdirplus use.

v2 of this patch incorporates Miklos's suggestion to simplify the code,
as well as improving consistency of macro names and documentation.

Signed-off-by: Eric Wong &lt;normalperson@yhbt.net&gt;
Signed-off-by: Miklos Szeredi &lt;mszeredi@suse.cz&gt;
</content>
</entry>
<entry>
<title>Synchronize fuse header with one used in library</title>
<updated>2013-02-07T10:58:12Z</updated>
<author>
<name>Miklos Szeredi</name>
<email>mszeredi@suse.cz</email>
</author>
<published>2013-02-07T10:58:12Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=7e98d53086d18c877cb44e9065219335184024de'/>
<id>urn:sha1:7e98d53086d18c877cb44e9065219335184024de</id>
<content type='text'>
The library one has provisions for use in *BSD, add them to the kernel one too.
They don't hurt and ease maintenance.

Signed-off-by: Miklos Szeredi &lt;mszeredi@suse.cz&gt;
</content>
</entry>
<entry>
<title>fuse: send poll events</title>
<updated>2013-02-04T15:14:32Z</updated>
<author>
<name>Enke Chen</name>
<email>enkechen@yahoo.com</email>
</author>
<published>2013-02-04T15:14:32Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=0415d291022543d83ee799e9ffee08d856bca6e8'/>
<id>urn:sha1:0415d291022543d83ee799e9ffee08d856bca6e8</id>
<content type='text'>
commit 626cf23660 "poll: add poll_requested_events()..." enabled us to send the
requested events to the filesystem.

Signed-off-by: Miklos Szeredi &lt;mszeredi@suse.cz&gt;
</content>
</entry>
<entry>
<title>fuse: bump version for READDIRPLUS</title>
<updated>2013-01-31T16:08:11Z</updated>
<author>
<name>Miklos Szeredi</name>
<email>mszeredi@suse.cz</email>
</author>
<published>2013-01-31T16:08:11Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=23c153e54197171f30b889d9654929d74b6599d5'/>
<id>urn:sha1:23c153e54197171f30b889d9654929d74b6599d5</id>
<content type='text'>
Yeah, we have a capability flag for this as well, so this is not strictly
necessary, but it doesn't hurt either.

Signed-off-by: Miklos Szeredi &lt;mszeredi@suse.cz&gt;
</content>
</entry>
<entry>
<title>fuse: implement NFS-like readdirplus support</title>
<updated>2013-01-24T15:21:25Z</updated>
<author>
<name>Anand V. Avati</name>
<email>avati@redhat.com</email>
</author>
<published>2012-08-19T12:53:23Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=0b05b18381eea98c9c9ada95629bf659a88c9374'/>
<id>urn:sha1:0b05b18381eea98c9c9ada95629bf659a88c9374</id>
<content type='text'>
This patch implements readdirplus support in FUSE, similar to NFS.
The payload returned in the readdirplus call contains
'fuse_entry_out' structure thereby providing all the necessary inputs
for 'faking' a lookup() operation on the spot.

If the dentry and inode already existed (for e.g. in a re-run of ls -l)
then just the inode attributes timeout and dentry timeout are refreshed.

With a simple client-&gt;network-&gt;server implementation of a FUSE based
filesystem, the following performance observations were made:

Test: Performing a filesystem crawl over 20,000 files with

sh# time ls -lR /mnt

Without readdirplus:
Run 1: 18.1s
Run 2: 16.0s
Run 3: 16.2s

With readdirplus:
Run 1: 4.1s
Run 2: 3.8s
Run 3: 3.8s

The performance improvement is significant as it avoided 20,000 upcalls
calls (lookup). Cache consistency is no worse than what already is.

Signed-off-by: Anand V. Avati &lt;avati@redhat.com&gt;
Signed-off-by: Miklos Szeredi &lt;mszeredi@suse.cz&gt;
</content>
</entry>
<entry>
<title>UAPI: (Scripted) Disintegrate include/linux</title>
<updated>2012-10-13T09:46:48Z</updated>
<author>
<name>David Howells</name>
<email>dhowells@redhat.com</email>
</author>
<published>2012-10-13T09:46:48Z</published>
<link rel='alternate' type='text/html' href='https://git.stealer.net/cgit.cgi/user/sven/linux.git/commit/?id=607ca46e97a1b6594b29647d98a32d545c24bdff'/>
<id>urn:sha1:607ca46e97a1b6594b29647d98a32d545c24bdff</id>
<content type='text'>
Signed-off-by: David Howells &lt;dhowells@redhat.com&gt;
Acked-by: Arnd Bergmann &lt;arnd@arndb.de&gt;
Acked-by: Thomas Gleixner &lt;tglx@linutronix.de&gt;
Acked-by: Michael Kerrisk &lt;mtk.manpages@gmail.com&gt;
Acked-by: Paul E. McKenney &lt;paulmck@linux.vnet.ibm.com&gt;
Acked-by: Dave Jones &lt;davej@redhat.com&gt;
</content>
</entry>
</feed>
